Primary hepatocellular carcinoma (PHC) is a common malignancy whose mode of oncogenesis is currently unknown. Preliminary studies have shown, however, genetic changes in tumors consistent with recessive events. These findings suggest that recessive oncogenic mechanisms may be important in PHC. The overall goal of this research is to ascertain genetic changes occurring in primary liver tumors. A molecular genetic karyotype will be derived for a collection of approximately 100 PHC tumors by typing RFLPs for polymorphic marker loci and candidate genes distributed throughout the human genome. This molecular genetic karyotype will permit evaluation of the genetic constitution of PHC tumor genomes not currently possible by standard cytogenetic techniques. Specifically, observed changes will be used to make a comprehensive test of the recessive model of oncogenesis in PHC. A systematic survey of candidate loci will permit evaluation of the significance of isolated reports of oncogene or growth factor changes in PHC. The information from the molecular genetic karyotypes will also be correlated with clinical variables to evaluate their joint significance in PHC. Patient data show wide variation in natural history and clinical presentation of disease. It is not known to what extent this variation reflects underlying genetic variation of the tumors. Observation of consistent cytogenetic differences in hematopoietic tumors has proven useful in classification and promises to be important in treatment. This study will attempt to determine if genetic factors are also of clinical importance in PHC.